Jb4 not working on 2015 640xi

[WWnyc]

Wow, amazing and highly educational.
Thanks for a great piece.

[chask]

Your welcome. It was my pleasure. In case you couldn't tell I really love this stuff.

Most of that post was based on work done from back in the 30's to around the 70's - not mine of course. A lot of it comes from high speed photography of the combustion process - that alone was pretty fascinating to see. Seeing the actual pictures of combustion really helped make it very easy to understand the basics of it. As I alluded to, the turbulence of the mixture plays a big part. That the air enters through the intake valve(s) on one side of the chamber causes some. The piston coming up on the compression stroke causes some more. The shape of the chamber also has a huge affect all of that. Then there is the location of the spark plug - wherever the engine designer managed to fit it in after getting the intake and exhaust valves in there. When the electrical arc occurs across the spark plug gap the mixture is flowing past it so quickly that the flame sort of makes a swirl at first. But that changes rather quickly.

Although there are some variations between cycles, the photo sequences I saw seemed to show that the end gas would most often occur in roughly the same area of the chamber. Often it was near the exhaust valve(s) - they tend to be particularly hot spots since some of the burning is still going on when they open. It all has to go past that valve(s) on the way out and there is not much to cool it. The exhaust valves on the S63TU motor are hollow and filled with sodium to facilitate some cooling. I think they are on most engines but don't count on that - I'm not a mechanical guy (electronics and control systems are my forte). But that valve and the area around it present hot spots that can help raise the tendency for self ignition or detonation in any of the end gas that happens to be around them. The intake valves get a fresh shot of relatively much cooler air every time they open so that side of the chamber stays a bit cooler.

Detonation is also far more likely when you raise the pressure of the mixture (in other words go into boost) and the temperature. The temperature of the intake air is also boost related. When the boost is really high the inefficiencies of compression can easily raise the intake air temperature by a couple of hundred degrees or more. Back when I was involved in this stuff running a bar of boost was really pushing it. Most kit makers and manufacturers were running 8-10 psi. Temperature increases were around 150-200 over ambient - on a 90 degree F day that would mean air temps of 240-290 degrees F. At a bar (14.7 psi) you could easily see 250-350 degree F increases over ambient. Getting the temperature down is why intercoolers are such a big deal. The really high air temperature is why they can use engine coolant (already at around 200 degrees F) as the cooling medium. It is also why injection of some water or a water/methanol mixture helps. The heat of vaporization reduces the temperature and the temperatures are so high that adding water does not hurt the power output at all. This cooling effect is also present when you run a fuel with alcohol in it (ethanol or methanol). The trick to making power is getting air in the engine. You can get more into the combustion chamber if it is colder.

I hope the OP and others can understand why it is important not to fool around with this stuff. Although I have not used their products, from what I have seen BMS and the others making similar systems have tried to work this out to provide a reasonable increase in performance without jeopardizing longevity very much. You can easily spoil those efforts by ignoring their recommendations. Although it could, it most likely won't go boom the very first time - you are more likely to see a gradual fatiguing of something before a catastrophic failure. How quickly that happens depends a lot on how hard you push it.

It also does not help to drive it really hard all the time. If you do you might want to think about doing other things that would help mitigate the effects of demanding that kind of performance out of your engine. Fortunately BMW tends to design things to last. That and the way they drive are the main reasons I love driving them. If you are going to push it, I'd suggest you pick components from people you trust are good at doing this and then listen to them. If you want to push it further, do it the right way. I think BMS recommends (has maps for) using race gas and octane booster and meth injection.

You can usually get away with it for a while. But if it fails, your car is going to be down for a while. Even if you opt to just put another stock motor back in there, these engines are not just lying around everywhere. It'll take some time to get shipped to you and then a little more to get it installed. Just something to think about before you push it too hard.

[Bönz]

Originally Posted by chask

Getting the temperature down is why intercoolers are such a big deal. The really high air temperature is why they can use engine coolant (already at around 200 degrees F) as the cooling medium.

I just read both of chask's posts and this gentleman is absolutely correct in every respect. This is why I have never had much confidence in aftermarket tunes (piggy-back or remaps). Even if they were to get everything right in terms of reading every sensor and controlling every parameter on these multidimensional maps, nothing is being done to increase the strength of the engine components. More power equals more stress. If you use it, it's only a matter of time. If you don't, why are you spending the money?

Regarding the quote above, one interesting note about the S63Tu is that it uses a completely independent cooling system with 3 radiators for the intercoolers and DMEs. This just reinforces the point how important it is to keep the charge air temperature down.

[yipyip99]

Originally Posted by chask

First off, a few disclosures are in order. I do not have a BMS tuner nor have I ever used one. I do not know how aggressive their tuning is - air/fuel ratios and ignition. I do not know how much boost they are running. I also am still learning how aggressive the stock BMW tuning is either. However, I have been around a whole lot of turbo motors (worked at a place that did this), some of them running unreal boost levels (even one that was running 29 psi boost). I've worked on some turbo engines capable of variable M fuel, neat gas up to M85 via a tank mounted sensor.

This stuff is not like a grenade where you pull the pin then there is a guaranteed explosion after a few seconds of operation. Oh, it certainly can be like that, but not always and more importantly with engines not usually. If (and I do mean if, because I really don't know exactly how much is being attempted by one of the BMS tuners at all, let alone about their specific maps - like Map 7) - if they are pushing the boost, ignition timing and A/F (air to fuel ratio) to the point they match a tune (or Map) to the particular characteristics of a fuel (like octane for gas or ethanol/methanol) then I would bet it is pretty tight on the tuning - they have pushed boost, timing and fuel to what they think the engine will reliably handle over time. I would think that is especially true if they have a different map for 93 average octane pump gas and 100-101 average octane "race" gas.

Assuming that is the case, what you are more likely going to see is a gradual weakening of one or more parts until enough stress fractures have been created that some day there is a catastrophic failure of a component or two. At that point it probably will go boom. Though it is in no way the same process, I would liken it to bending a piece of solid core wire back and forth to friction melt it: grab a piece of solid core wire in both hands. whip your wrists back and forth as rapidly as you can, and it will soon break apart. The wire is going to hold up for a while and everything will seem fine. But eventually after it has been worked beyond its design limits long enough the wire weakens. You will gradually feel it when it starts to approach this point as it will suddenly become noticeably easier to bend the wire. Then it will fail and break in to two pieces. The problem with an engine is you don't know when it is getting to this point. There are rarely any signals until it just goes. When it does, most of the ones I have seen usually take out enough other stuff that it's hard to tell what happened first.

This is a very simplistic explanation of detonation and what it does. In reality there are a lot of other variables. Turbulence of the mixture in the combustion chamber is a big one. But it will give you an idea of what is going on and why you might want to rethink running on map 7 unless you put some 101 octane and/or meth in it

When your spark plug ignites the fuel around it, everything does not burn all at once. There is a flame front that progresses across the combustion chamber. As I recall from when I used to work with this about 25 years ago the flame front velocity is about 200-300 feet per sec (or maybe it's 200-400 fps). As the flame front progresses the hot burnt gases behind it are rapidly expanding so it is further compressing the end gas (the remaining unburned fuel). This brings it closer to self ignition. If all goes well that never happens and all the fuel burns at a controlled rate. The complete burn takes a few milliseconds. What you want to have happen is for the pressure from the burn to hit and push on the top of the piston when it will produce the greatest force on the crank - the most torque, and since HP = Torque * RPM/5252 the most horsepower too. This burn rate is why ignition timing varies with engine speed. That the burn rate changes with temperature and pressure (and other factors) is why they vary timing with those variables too.

Knock occurs when the conditions of the unburned fuel in the combustion chamber reach the point of self ignition. The worst case is dieseling - where there is no ignition and the mixture just blows up under compression, just like a diesel engine uses compression to ignite the mixture. Most often in a gasoline motor it is some small remnants of unburned fuel that go off on their own rather than a controlled burn.

When detonation occurs, the end gas goes off all at once; it essentially explodes. This explosion produces a much larger pressure. It also produces a concussion wave that progresses across the combustion chamber at around 2,000 fps. Unlike the flame front that dies when all the gas is gone, this pressure wave only dissipates by friction with the sides of the combustion chamber and repeated collisions with the top of the piston and the cylinder head (and I guess after the exhaust valve has bled some of it out of the combustion chamber into the exhaust system). This is the "knock" you hear. That friction of the pressure wave running along the walls and hitting the piston and cylinder head makes things very hot. Thus the fuel that comes into the chamber on the next intake cycle will be closer to the point of self ignition too. If allowed to continue for very long this heat creates extra and perhaps excessive heat stress on the engine. This is not too good either - no, sarcasm aside, this can be very bad. Look at all the radiators BMW puts on this engine to keep it cool and you will get an idea how important running cool is. Knock is also literally hammering the engine. Though I am an electrical type, I know that impulse loading of anything mechanical is usually not good, especially when it is very hot (materials are usually weaker when heated). In case you are not aware, a big part of the reason that the A/F drops to 12-13 range as you approach wide open throttle is that the extra fuel helps cool the combustion chamber.

All this said, an engine will make a little more horsepower when it knocks just a little (I'll be glad to explain that if anyone wants to know). Ideally for the best efficiency and a little more power you want to have a little knock, but very little. But since knock usually generates extra heat and heat increases the chance of uncontrolled knock, running on the threshold of very light knock is not something that is easily maintained. That is especially true when you consider how large the cyclic variations between successive engine cycles can be - in other words, the amount of fuel and ignition timing and air entering the cylinder is not exactly the same every time. They have gotten a whole lot better at controlling it in the last 25-30 years but there is still some variation. When you are trying to control something with precision, the last thing you want is another variable to worry about. The extra heat from detonation is about the worst thing that can happen.

Also consider that the engine control system will try to adapt to the knock it is seeing. I do not know the actual strategy the DME employs but generally reducing ignition timing and boost are a good first guess. I'd go with timing first (since it is the quickest to control -you can get the very next cycle), then boost (probably at least a few revs if not a bunch will occur before that can be reduced), and since it is drive by wire maybe throttle too if the other two don't bring the knock under control. Just like your car's engine computer will adapt to the changes you make to improve performance (bolt on stuff and a tune) it will also try to adapt its tuning over time to reduce the knock. It will succeed up to a point, but from what I understand about the BMS system (the fact that they have maps for various operating conditions) I would guess that it cannot completely adapt enough to completely mitigate the problem.

Look at Formula One for an example of what happens. Those motors are stressed like crazy. The tolerances are so tight that they have to run hot water through the cooling system and hot oil through the lubrication system to bring the mechanical components up to operating temperatures (or closer than just sitting there at room temp) before they can even try to start the engine. Even with all that precision and all the finest components unlimited sums of money can buy these engines are still pushed so hard that they have to use 4 or 5 engines to make it through the entire 20 or so races that comprise a season. What you are doing when you run an engine in knock all the time is heavily stressing it.

I suspect that this is what you may be doing to your engine by running it on a map that is intended for race gas - pushing it too hard, essentially beating on it. Think running with a quart or two less oil than it needs to be full, or straight water instead of a coolant/water mix, or not enough water. You may get lucky and it will last until you move on to another car. Or it might go the net time you floor it. Frankly, with an engine that runs about $30-40K I would not want to chance this. But maybe you have the resources and can be without the car while it is being rebuilt, or a new one acquired and installed. That definitely would be a great time to really hop up that motor .

I have heard the rods are the weak point. I also read in the S63TU Engine Manual (appears to be a BMW publication that you can find it in this forum) that the pistons are cast. They are Mahle's (very good pistons from a great piston maker - at least they were back when I was seriously in this stuff) but they are cast and not forged. On top of that, the mechanical compression ratio is 10.0:1. That is pretty high for a turbo motor, and especially one running around a bar of boost (14.7 psi). From what I have read these engines sometimes produce even more boost. I am not sure but I think the BMS race gas tune may be up closer to the 18-20 range. In their forum I have read 20-22 or so but I may have that wrong - I'm still trying to figure out all the BMS and BMW nomenclature). If this is the case, you are asking for trouble.

A lot of this will depend on your driving style too. If you do just an occasional hard acceleration to get on the highway or pass someone you are more likely to get away with it for a longer time. If you are doing banzai autobahn runs (high speed driving and repeated accelerations at those speeds) , sustained and/or repeated hard accelerations, or any kind of really hard driving where you are using most of the engine performance and the engine is not given a chance to cool down you are more likely to run out of luck much sooner - think F1 engines. Think ticking time bomb.

It's your car and your choice, but I wouldn't do it. Not for very long anyway.

On the other hand -
If you do end up with a lunched motor and you want to rebuild (assuming the critical things are still intact - not sure it's likely with an Alusil lined aluminum block but who knows), but this sure would be the perfect time to go for a huge power motor or even just beef up the one going in with forged pistons and rods. Even if you have to start with a completely new engine, that new engine would still be out of the car and it'd be much easier to build a really super strong and high power one then, rather than having to pull it out of the car.

I have always thought BMW engines were heavily overdesigned. This has been reinforced by both my personal experience and observations of others' experiences with them. For example when they went into F1 I read they used their stock 4 cylinder block and got something like 1000 hp out of it. Wikipedia says 600 out of a 1.5 liter with 2.8 bar boost (41 psi) but I seem to recall C&D or R&T reporting about 1000 one year - not something a hard core car nut driving a 4 cylinder Bimmer forgets. Still 600 out of a 1.5 is 400 bhp per liter. I don't see why you couldn't get a very reliable 250-300 bhp per liter out of this engine - maybe more with all the high tech goodies on it, perhaps a lot more. I have seen that or perhaps a bit more on a car with very basic electronic fuel injection and a very early distributorless ignition system (mapped RPM and MAP) that ran 70K miles before I lost track of it (still running strong). It did not have double VANOS VVT, or scrolled turbos with a pulse manifold (360 degree opposed cylinders feeding the same scroll and opposing cylinders feeding the other scroll) - I suspect they are probably roller bearing too. To do this at the least it would take forged pistons, forged Ti rods and a couple of bigger turbos; probably some head work too (port, polish and match to manifolds), maybe cams as well. I have read the DCT would probably need some help to handle all that but hopefully just some clutches and a maybe few other easy components (I know next to nothing about transmissions - no, less than nothing). And a whole lot of computer tuning. Most likely it would be a very expensive venture but think of the results! You'd definitely need a whole car wide body kit to cover the rubber it would take to get that kind of power on the ground. That or a new set or rear tires every week or two - maybe every day or two. Talk about a beast.

Thanks for the advice.

It's a brand new car, so all components should be able to handle it for a bit since I don't do any hgwy driving during the week, and my commute to work is only 4miles.

However, I don't want to take the chances of blowing the engine, so I will keep it on map 5 instead.

My car is a lease, and I won't buy it at the end of the lease. I am waiting for the 2017 M6 to come out and have a high power car from stock.